172 
Journal of Agricultural Research 
Vol. XXVI, No. 4 
It is interesting to compare the monthly percentages of otocephali 
with the monthly averages in size of litter and in percentage of the 
normal young raised to weaning, among the young from the same mat- 
ings. Size of litter and mortality before weaning are characters which are 
known to be affected to a marked extent by seasonal conditions. The 
monthly averages for size of litter, percentage of young not otocephalic, 
and percentage of normals raised to weaning are shown graphically in 
figure 3. In order to smooth out the irregularities due to small numbers, 
the figures for each month were combined with those for the preceding 
and the following month and averaged. These smoothed averages are 
shown by the heavy fines. The three curves show considerable similarity. 
In order to bring this out more clearly, however, it is necessary to com¬ 
pare the percentage not otocephalic with the size of fitter about half a 
month later and the percentage raised of the preceding month. The 
curves are shoved over in this way in figure 3. Such relations are 
not unexpected. Size of fitter is presumably largely determined by con¬ 
ditions two to three months before the fitter is born, the gestation period 
beihg about two months and a week. Development as an otocephalus 
must be determined very early in development, probably before the 
appearance of the medullary plate, from the experiments of Stockard 
and others. Assuming that the condition of the dam has an influence, 
poor conditions should result in otocephali a little earlier than in small 
litters, but an effect on the mortality of the young should appear still 
earlier. The conditions during the month preceding birth are probably 
most important in this connection, their cumulative effect determining 
the mortality at birth and between birth and weaning. Most of those 
which die before weaning are unthrifty from the first. 
The correspondence between the monthly fluctuations shown in the 
figure, after allowing for a reasonable lag in each case, adds considerably 
to the evidence that seasonal conditions play a part, if not a very great 
one, in determining the occurrence of otocephali. 
SIZE OF utter 
As just pointed out, size of litter is to some extent an indicator of fav¬ 
orable or unfavorable conditions at the time of conception and shortly 
before. We have just seen that smallness of fitter and percentage of 
otocephali show parallel seasonal fluctuations with a lag of only about 
half a month. It is interesting to make a direct comparison between the 
size of the fitters in which otocephali were born with the size of those in 
which normals were born. Data are given in Table VI for otocephalus- 
producing matings. The number of otocephali and the number of nor¬ 
mals in each size of fitter is shown, together with an average. This 
average gives the size of fitter relative to individuals, and so is con¬ 
siderably larger than when the litter is taken as the unit, as in figure 3. 
Both in Family 13 and in the other stocks otocephali were born in smaller 
fitters than normals, and in the former case the difference is 4.7 times its 
probable error, although only 1.5 times in the latter. In both combined 
there is a difference of 4.4 times the probable error. There can be little 
question that conditions which determine small fitters are relatively likely 
to cause otocephalic development. The importance of such conditions, 
however, is not great. 
